New Trading Systems and Methods, Fourth Edition

CHAPTER 8

Time-Based Trend Systems

The previous chapters developed the tools for calculating trends—a traditional moving average, weighted averages, exponential smoothing, and regression. To profit from identifying the trend requires the use of trading rules and the selection of specific parameters that define the trend speed and an acceptable level of risk, among other factors. This chapter first discusses those rules that are necessary to all trading strategies, then gives examples of actual systems. The selection of trend speed is handled only briefly here but is continued with a detailed analysis of these and other systems throughout the book, and especially in Chapter 21. Towards the end of this chapter, a section on viewing sequences of trends may prove interesting to the analyst trying to be innovative.

WHY TREND SYSTEMS WORK

Trend analysis is the basis for many successful trading programs. Whether these programs are trend following, taking positions in the direction of the trend, or mean reverting, taking positions contrary to price movement, identification of the trend is an essential component. Trend systems work because

• Long-term trends capture large price moves caused by fundamental factors, primarily initiated by government policy and reflected in interest rates, foreign exchange, and the balance of trade.
• Prices are not normally distributed but have a fat tail. The fat tail means that there are an unusually large number of directional price moves that are longer ...